Sea Ice Modeling and Data Assimilation (SIMDA)

Integrated Foundations of Sensing, Modeling, and Data Assimilation for Sea Ice Prediction
Department of Defense Multidisciplinary University Research Initiatives program through the Office of Naval Research

Empirical bayesian modeling of a complex Valued scene using joing sparsity

17 Oct 2022 - Dylan Green


Using our multiple measurement vector approach with an empirically-derived mask in the magnitude prior, we sample the posterior distribution of a complex-valued signal. Here, using four measurement vectors with noise added in the Fourier domain, we show the magnitude of an example image (top left) recovered in three ways: using just the inverse Fourier transform of the mean of the data (top right), finding the mean of the magnitude when a nonempirical prior is used (bottom left), and finding the mean of the magnitude when the empirical prior is used (bottom right).